Two phase PSV sizing - (vaporization before discharge point) 5

[SF6-146]

Dear all,

We know the HEM method to sizing the valve but we have no idea how to size a valve that arrive fluid in two phase... for example, there is a fire and the reactor jacket is heating... so the fluid is vaporizing and arrive in two phase to the PSV...

I want to ask if there is any experimental paper or some information to predict what would happen, or if i can assume the equilibrium.. etc..+

Thank you in advance,

 

[PaoloPemi]

for vapor+liquid you may wish to estimate pressure drop in inlet piping otherwise the procedure is the same,
if you accept the limits of HEM model (homogeneous equilibrium and vapor + liquid traveling at same speed) use HEM model,
the scope being to estimate the max flux (critical flow / sonic velocity) at the throat condition.
There are several methods applicable, the software which I utilize (Prode Properties) calculates directly speed of sound / critical flow (for HEM and HNE models) but you may find different methods in literature.

 

[TiCl4]

API 520, Part I, Annex C Sizing for Two-phase Liquid/Vapor Relief contains PSV sizing equations for flashing fluids. These equations are based on HEM, and assume, in their words, that "thermal and mechanical equilibrium exist as the two-phase fluid passes through the PRV".

I suggest you start there for sizing PSVs for flashing flow.

Edit: Jacket relief sizing for fire case is rife with complications. Due to the high surface area to volume ratio, required relief sizes are often in excess of the available nozzle size on the jacket. Location in the jacket matters, as relief from the bottom will see flashing liquid during 99% of the relief event while relief from the top will see incoming two-phase flow.

The best means of dealing with an external fire case on the jacket is to prevent the jacket from being blocked-in during a fire. There are a few fire-safe ways of doing this - all rely upon the heat from the fire to melt a component that renders the system safe.

Reference other threads in this forum for guidance.

 

[SF6-146]

hello to all,

I fount the ISO 4126 part 10 that explain very well all the cases.

I put here the link!

https://qdoc.tips/bs-iso-4126-safety-valves-specs-part-10-pdf-free.html

Thanks to all!

 

[don1980]

Senrof - TiCl4 pointed out a very important fact about protecting vessel jackets from fire exposure, and I suggest you re-read that before proceeding. Due to the high surface-to-volume ratio, the liquid in the jacket will heat up and vaporize very quickly. And, since there's no opportunity for liquid-vapor disengagement, the vapor will sweep liquid into the inlet of the PSV (thus, your correct assessment that this is a two-phase design case). But before proceeding, sit back and consider the following:

1) As stated by TiCl4, you'll probably find that the PSV requires a nozzle that's larger than anything currently available on the jacket. Furthermore, you'll likely find that the required nozzle is actually larger than the jacket itself.
2) Even if you design a PSV for two-phase relief (and are able to install it), recognize that this PSV will empty the liquid contents very quickly.
3) Then, the jacket walls rapidly heat up and fail.

So, one should ask, what has been accomplished by sizing this PSV for the fire scenario. The answer is, nothing is really accomplished. The jacket fails regardless of whether or not you size the PSV for fire. Accept/recognize the fact that a PSV is incapable of providing the jacket any meaningful protection from fire exposure, and focus your attention on other protective measures which are at least partially effective (e.g. NFPA 15 water spray and fire-resistant insulation).

Note that I'm not suggesting you omit a PSV from the jacket - one is needed for compliance. Instead, I'm just saying that there's no justification for sizing that jacket PSV for a fire scenario. Size it for thermal expansion, or for some other scenario, if there are other credible scenarios.